In the previous funding period we identified areas of the brain essential for distinct cognitive processes underlying lexical tasks, by identifying areas of tissue dysfunction associated with specific lexical deficits immediately after left hemisphere ischemic stroke in 317 patients. We used MRI scans, including diffusion weighted imaging and perfusion weighted imaging, to identify dysfunctional tissue and language tests to identify deficits in lexical processing within 24 hours of stroke onset. We also determined which deficits recovered in association with tissue recovery in specific regions by Day 3-5. We found new evidence that lexical tasks such as naming, reading, and spelling depend on overlapping networks of brain regions, but that distinct cognitive processes underlying these tasks depend differentially on one or more of the neural regions in this network. In this competitive continuation we propose to use this methodology to investigate the neural regions critical for a number of language and cognitive processes that all have been hypothesized to rely on Broca's area, or have been reported to be impaired in patients with Broca's aphasia, or both. We will test specific hypotheses about cognitive processes that depend on Broca's area (based on functional imaging and chronic lesion studies), and hypotheses about cognitive/linguistic deficits underlying Broca's aphasia (generated from psycholinguistic and neuropsychological studies). We aim to determine what cognitive functions underlying language are specifically associated with acute damage/dysfunction of Broca's area and/or chronic damage to Broca's area. In addition, we aim to determine whether or not impairments of these functions are associated with the clinical diagnosis of Broca's aphasia, or components of this clinical syndrome, in the acute or chronic stage. This research is expected to clarify the role of Broca's area in language and cognitive processing, and to clarify the relationship between deficits in cognitive functions attributed to Broca's area and the language characteristics of Broca's aphasia (e.g. agrammatic speech, apraxia of speech, and asyntactic comprehension). We expect the results to contribute to the understanding of the neural and cognitive mechanisms underlying sentence comprehension and production. The results will also have clinical impact by aiding in functional prognosis in acute stroke and in identifying appropriate candidates for intervention to restore blood flow in patients with language deficits due to acute stroke.